A hypergradient-free social-gradient flow is proven to be a descent direction for the planner's objective and converges to the unique socially optimal incentive when social cost depends on agents' joint actions.
Adaptive Incen- tive Design with Learning Agents
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.OC 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
RAID algorithm achieves asymptotically minimal regret in incentive design under information asymmetry via a switching policy and a strongly consistent least-squares type estimator that relaxes persistence-of-excitation assumptions.
citing papers explorer
-
Incentive Design without Hypergradients: A Social-Gradient Method
A hypergradient-free social-gradient flow is proven to be a descent direction for the planner's objective and converges to the unique socially optimal incentive when social cost depends on agents' joint actions.
-
Adaptive Incentive Design with Regret Minimization
RAID algorithm achieves asymptotically minimal regret in incentive design under information asymmetry via a switching policy and a strongly consistent least-squares type estimator that relaxes persistence-of-excitation assumptions.